The use of chemical indicators to source human, animal and bird pollution of Fraser Island lakes and streams.

Page created by Jeremy Cobb
 
CONTINUE READING
The use of chemical indicators to source human, animal and bird
              pollution of Fraser Island lakes and streams.

                  Neil Tindale, Daryle Sullivan and Peter Brooks
                 Faculty of Science, Health and Education (FSHE)
                         University of the Sunshine Coast
                      Maroochydore DC, QLD 4558 Australia

Abstract:
The presentation covers the use of a chemical indicator technique to investigate
whether tourists are polluting the fresh water ecosystem s on the relatively pristine
Fraser Island. T he objectives to the development and use of the technique included:
      Seeking to quantify and distinguish faecal pollution inputs from human and
          other, natural animal sources.
      Identify lakes and creeks on Fraser Island that are impacted by human
          visitation.

The focus of this paper is on the methodology and use of the chemical indicator
technique. Samples of surface waters and lake floor sediments were collected from
Fraser Island and analysed by GC-MS. The analytical technique is based on using
faecal sterols as indicators of faecal pollution. The results showed that the sterols
were very low to low in the lake sediments and undetectable in the surface water
samples. Ratios of sterols indicated that water fowl are the l i kely predominate faecal
source. Human input appeared to be negligible compared to natural sources.

Introduction:
The rese arch investigated whether tourist visitation of the relatively pristine Fraser
I sland lakes is polluting the fresh water ecosystems. The other aim of the study was
to develop and test the faecal sterols chemical indicators methodology for use at
USC. This study was dependant on access to a newly purchased gas
chromatograph -mass spectrometer (GC-MS) in the FoSHE research instruments
laboratory. If the study was successful, it would develop analytical capability in
organic chemical contamination that would be useful to the environmental science
and public health program s within USC and the region. While microbial techniques
can identify the presence or absence of faecal microbes, then generally don’t provide
much detail on relative types and amounts of sources.

Objectives and workshop structure:
The aim is to introduce a relatively little used chemical indicator technique which can
be applied to local/regional water quality studies. It is particularly useful in
determining the likely sources of faecal contamination of surface and subsurface
waters and sediments, an d potentially capable of di stinguishing relative inputs of
faecal material. Effectively it duplicate s and applies the faecal sterol chemical
indicators technique that has been used in CSIRO (Leeming et al., 1996, Leeming et
al., 1998 ) and on lake sediments (Logan et al., 2001), river water (Suprihatin et al.,
2003) and groundwater (Roser et al., 2002) within Australia. This technique was also
u sed in a much larger scale by the US Geological Survey to study organic
contamination of streams across 30 states in the continental USA (Kolphin et al.,
2002). While the technique was initially developed over a decade ago now, it is still
only used relatively sparsely as it requires careful attention to trace analyses and the
interpretation of results is often difficult. Detection limits for the faecal sterols are
required be down in the parts-per-trillion level and interpretation can be confusing
due to variability in pollutant and natural sterol levels.
                                       nd
                                      2 Queensland Coastal Conference, Gold Coast, May 2009
Method:
Samples of surface waters and lake floor sediments were collected from Fraser
I sland during two field sampling trips to Fraser Island in the summer of 2005.
Samples were returned to USC and processed prior to analysis by GC-MS. The
analytical technique is based on using faecal sterols as chemical indicators of faecal
pollution. Sterols in the diet are biohydrogenated by anaerobic bacterial to a mixture
of stanols. The unique ratios of sterols and stanols in faeces, arising through diet and
metabolism in each animal species, can be used to track the source s of faecal
contamination in water bodies and sediments. The faecal sterols are very surface
active chemicals and bond onto sediment particles so are usually found in the
sediments or suspended sediments. Their lifetime dissolved in water is short (~ 1
week) compared to months in anaerobic sediments.

Sediment samples were collected f rom selected lakes and streams in glass bottles,
and transpo rted to the University on ice. They were stored at 4ºC and processed and
analysed within 24 hours of collection. Processing included draining of excess water
by vacuum filtration. Then each sedim ent sample was weighed and 50mL of hexane
and 50mL of methanol added, together with one gram of sodium carbonate and
cholestane internal standard. The samples were then tumbled for 24 hours. The
mixtures were vacuum filtered on pre-wa shed filters, and the filtrate transferred to
separating funnels. The lower, aqueous methanol layer was drained away. The
hexane layer was washed with water and dried over sodium sulphate. Then the
samples were evaporated under nitrogen gas and placed into a freeze drier for one
hour at less then 1 milli-bar. The dried samples were then re-dissolved with 1 mL of
hexane. To improve the volatility and stability of sterols in the extract and to improve
detection limits on the GC-MS, they were converted to their O-trimethylsilyl
derivatives prior to analysis through the addition of 20µL trimethylsilyl imidazole.
The samples were allowed to stand for three hours before being analysed by GC-
MS.

Water samples we re collected in 10-liter high -density polyethylene (HDPE)
containe rs. The containers were rinsed 3 times with water from ~1 5 cm below the
surface of each site before the samples were collected. The samples were then
transported to USC and placed into cold storage at 4˚C, and processed and analysed
within 36 hours. Each sam ple was filtered through 142mm/0.45micron glass fibre
filter. The sterol -based lipids were extracted from the particulate matter trapped on
the glass filters using a sim ilar extraction process as was used above with the
sediments.

Sample extracts were analysed on a Varian 3900 gas chromatograph (GC) equipped
with a ZB-5ms-30m x 0.25mm capillary column. Technical details: Initial injector
temperature was set at 320˚C and the split ratio was initially 50:1 and closed for 30
seconds on injection. The constant column gas flow was 1.0mL/min. The oven
temperature was programmed at 200˚C on injection and increased at 20˚C/min to
240 ˚C, then increased at 3.0˚C/min to 320˚C and held for 5min. The total method run
time was 33.67 min. The GC was coupled to a Varian Saturn 2100T mass
spectrometer with a transfer line temperature of 310˚C, and compound ionization by
70eV electron impact. The positive fragment ions were analysed over a mass range
200 to 550m/z. The sterols /stanols of interest were identified by retention time
against known standards and by their mass spectra. Sterol/stanol standards of
cholestane, coprostanol, cholesterol, stigmasterol, sitosterol and sitostanol, were all
purchased from Sigma Aldrich, Castle Hill NSW. The 5α-ch olestanol and
epicopro stanol were purchased from Steraloids inc, Newport Road Island USA.
                                      nd
                                     2 Queensland Coastal Conference, Gold Coast, May 2009
Validation of the methodology was through the use of spiked samples. Known
amounts of some of the standards were added to sediment and water samples and
processed and extracted using the same techniques as above. The results from the
spi ked samples were compared to the unspiked.

An example GC-MS spectra for a sample of Lake McKenzie sediment is included
below:

Results and discussion:
We were able to detect the sterol and stanol compounds at sufficient sensitivity and
resolution to quantify the amounts and to distinguish the sources of these
compounds in samples of lake sediments, but not in water samples. Results for the
sampling on Fraser Island on December 5, 2005 are shown together with d ata, also
collected by us, for the North Maroochy river in the Sunshine Coast.

            Lake           Lake             Lake    Wabby      Ocean       Septic      Nth
            McKenzie       Birrabeen        Allam   Lake                   System      Maroochy
Coprostanol 0.4            0                2.4     1          1           6000        231
Cholesterol 229            317              233     118        204         100000      151
5a-         51             0                25      0          0           5300        74
Cholestanol
24-Et.      0              0                2       0          0           29000       96
Coprostanol
Sitosterol  793            413              3319    2105       1881        22000       74
Sitostanol  0              418              383     239        406         2000        18

Ratio         0.35         0.38             0.07    0.05       0.09        3.1         2.4
C27/C29
                                       nd
                                    2 Queensland Coastal Conference, Gold Coast, May 2009
Ratio 5b/5a   0.007         0               0.01     0.003       0.003       12.2         3.5

              Septic        Seagull         Duck     Swan        Possum      Roo          Dog
Ratio         3.1           3.7             0.6      0.25        0.4         0.5          4.9
C27/C29
Ratio 5b/5a   12.2          6.9             0.8      0           13          3.6          0.6
Concentration units: ng/g

The data in the lower part of the table are from the previous studies in Australia
(Leeming et al.,1998). This data shows that human faeces (septic systems) are high
in coprostanol. Coprostanol is also the major sterol in pig faeces. Cattle, kangaroo,
and possum s also contain coprostanol in their faeces, but at far lower concentrations
than humans. Leeming et al. (1998) note al so that while not present in human
faeces, epicoprostanol is generated in anaerobic septic and sewerage systems, and
can thus be a marker of aged human sewage contamination. Cattle, kangaroo and
possum herbivores had much lower C27:C29 ratios compared to humans. Bi rd
species exhibit a broad dietary intake, but Leeming e t al., 1998 and Walker et al.,
1982 noted low 5α stanols, including coprostanol in a range of bird faeces. The 5β
stanols were below detection in dog faeces. Invertebrates reportedly do not exhibit
coprostanol in their faeces (Leeming et al., 1998), similarly for fish and reptiles.

Data from Fraser Island showed that the sterols and stanols were very low to low in
the lake sediments and undetectable in the surface water samples. Ratios of C27 to
C29 sterols, C27/C29, and that of 5β to 5α st erol s, 5β/5α , indicated that water fowl
are the l i kely predominate faecal source. Human input a ppeared to be negligible
compared to natural sources on Fraser Island. In contrast, samples and data from
the North Maroochy river on the Sunshine Coast gave higher levels of stanols and
indicated that human septic emissions were p robably contributing to the elevated
levels and ratios.

Acknowledgements:
Grateful thanks to Dr Rhys Leeming for his advice and the loan of his filtration
equipment. This project was funded by the University of the Sunshine Coast through
a USC-Internal Research Grant

References:
Kolphin, D.W, E.T. Furlong, M.T. Meyer, E.M. Thurman, S.D. Zaugg, L.B. Barber and
H.T . Buxton, 2002. Pharmaceuticals, hormones and other organic wastewater
contaminants in U.S. streams, 1999-2000: A national reconnaissance. Environ. Sci.
Technol. 36: 1202-1211.

Leeming, R., Ball, A., Ashbolt, N., Nichols, P., 1996. Using faecal sterols from
humans and animals to distinguish faecal pollution in receiving waters. Water
Research 30, 2893 - 2900.

Leeming, R., Nichols, P.D., Ashbolt, N. 1998. Distinguishing sources of faecal
pollution in Au stralian inland and coastal waters using sterol biomarkers and
microbial faecal indicators. CSIRO Report 98-WSAA and Water Services Association
of Australia Report No. 204. Report prepared for the Water Services Association of
Australia Nov. 1998.

Logan, G.A., Fredericks, D.I., Smith, C. and Heggie, D.T., 2001. Sources of organic
matter in Wallis Lake. AGSO Research Newsletter, May 2001: 15-20, 2001.
                                       nd
                                      2 Queensland Coastal Conference, Gold Coast, May 2009
Roser, D., N. Ashbolt, M. Storey, R. Leeming, P. Scott, R. Bennetts, D. Ryken-Ra pp
and S. Toze, 2002. Application of physico -chemical parameters and particle-bound
biomarkers to indicate microbial contamination of aquifers. Water Supply 2(5-6 ):
419– 426.

Suprihatin, I., H. Fallowfield, R. Bentham and N. Cromar, 2003. Determination of
faecal pollutants in Torrens and Patawalonga catchment waters in South Australia
u sing faecal sterols. Water Science and Technology 47(7-8): 283–289, 2003.

Walker, R. W., C. K. Wm, W. Litshy, Coprostanol as an indicator of faecal pollution,
CRC Critical Reviews in Environmental Control 1982, 12, 91

                                      nd
                                    2 Queensland Coastal Conference, Gold Coast, May 2009
You can also read